Michelle Russell-Dowe, Head of Securitised Credit, Schroders

Securitisation and the Willingness to Repay

Securitisation was almost thrown out with the bathwater in the aftermath of the global financial crisis, as mortgage-backed securities (MBS) and collateralised debt obligations (CDO), containing sub-prime mortgages, almost brought down the financial system.

But since then the securitisation market has undergone significant change, both in regulations and attitudes towards risk.

And in a world where both equities and bonds look expensive, investors have become increasingly willing to revisit these structures.

Michelle Russell-Dowe, the Head of Securitised Credit at Schroders in New York and a veteran in this space, says many investors have come to the realisation that in contrast to securities that contained sub-prime mortgages, securities with more senior tranches held up relatively well during the crisis.

“The vast majority of what I would call securitised credit, or non-guaranteed securitised bonds that have a credit risk premium, was really beginning to be issued in the midst of the early-1990s,” Russell-Dowe says in an interview with [i3] Insights.

“Between then and the financial crisis, there wasn’t a really massively correlated event to test these structures. So you had structured finance and all of that was rated, but beyond the assessment of the rating agencies no one had proof of what the impact would be on the junior classes.

“I think the first thing that people took away from the financial crisis was that this was obviously a very critical test case for the concept of securitisation, or structuring, and I think what people found is that we can [do it].

“Many sectors, whether that was US consumer asset-backed securities (ABS), collateralised loan obligations (CLO), European MBS, or even senior commercial MBS, the senior securities in these structures generally were adequately protected in this type of recession, as was evidenced by very limited losses and downgrades.

“Where you saw the problems, obviously in the financial crisis, were where people had put more junior exposures together into the capital stack, in particular in asset classes where the underwriting had gotten quite weak, such as US mortgages.

“And so, I think, people acknowledge that there were problems, but it wasn’t with every securitised asset and then the problems that we saw in US mortgage securities have been principally addressed by regulation.”

But the crisis did cause investors to change their attitude towards risk and this led to a change in the use of valuation metrics to assess structured credit.

For example, the way in which mortgages are underwritten in the US differs from those in the United Kingdom or in Europe, where mortgage defaults affect the lender personally and don’t relate just to the property.

This difference caused some investors to focus on the wrong metrics in the past, Russell-Dowe argues.

“In the UK and Europe, generally a lender has personal recourse to a borrower. In the US, it’s much less frequent that this would be the case. [But] you can’t really change the fabric of the lending market,” she says.

“But what you can change is the idea that, principally, securitisation should be the study of a borrower’s ability and willingness to repay and only secondarily of the collateral value.

“And I think prior to the financial crisis people relied far too much on the collateral value and the idea that something like a home was important to a borrower.

“I think what people misunderstood was that by the time we got to the mid-2000s most borrowers actually considered their home as a part of their investment portfolio rather than as an economic good that they consumed, and that’s a pretty substantial change.”

The shift away from a focus on collateral and towards a borrower’s ability to repay was helped by new regulations under the Dodd-Frank Act.

“Once you get outside of a qualifying mortgage, the originator must use a standard called ability to repay (ATR) or they may not be protected when a borrower defaults,” Russell-Dowe says.

“That’s just another way of saying that now underwriting must focus on a borrower’s income relative to the payments that they must make on the loan, and I would argue that that’s the principal criterion upon which we should be judging the credit quality of the loans.

“And so regulation in the form of the qualifying mortgage limitations, or requirements for skin in the game, has been a real game changer across these markets.”

But to properly assess the credit quality of a loan, investors should not just focus on a single income figure, but delve deeper into how their income might be affected.

“It’s the idea of not just assessing what is the base case, but rather the potential for variation around it,” Russell-Dowe says.

“I think for those who are experienced in the securitisation market, the methods of analysis have changed fairly dramatically to assign a wider variance around the potential outcomes and to assessing the structure, or protection, to mitigate against that variation.

“So that’s all a lot of fairly nerdy language for saying that you need to really understand how the security risk can change over a wide range of scenarios. That should actually be the cornerstone of how you look at value in these securities.”

Assessing the value of securitised assets also means you need to have a good understanding of the recovery rate in case of default, something most rating agencies tend not to look at, she says.

Principally, securitisation should be the study of a borrower's ability and willingness to repay and only secondarily of the collateral value

“If you look at the rating agencies, and this is a generalisation, it might not apply to all of them, but the way that they assess risk is the risk of a dollar of principal loss to a given security, that’s a probability of default assessment,” she says.

“Where they’re weaker is in assigning a loss given default. If you have a default on a security, do you know what ultimately your recovery is? Because I think when you deal in more standard debt, such as a corporate or sovereign, you’re not dealing with the same sort of capital structure nuance that you are dealing with in structured finance.

“Especially if you have classes that are small relative to the total debt outstanding, you have a leveraged sensitivity to changes and that can dramatically move around your recovery potential.

“So I think that most people within these markets have recognised that to be dependent upon the rating agencies as the arbiter of credit in the first case is probably not great.

“But also rating agencies aren’t in the business of generating portfolio returns; they’re in the business of saying good or bad from a probability-of-default perspective.”

The shift in attitude towards risk and valuation metrics has caused the premium for holding these instruments to increase, she says.

“Before the financial crisis, the risk premium that you would receive for owning an AAA-rated CLO was probably less than 50 basis points over LIBOR and today that same number is closer to a 120 basis points,” she says.

“Because of components like this idea of the impact of variance on security performance or that there is more regulation around who can buy the securities, you’ve seen a fairly durable repricing of the risk, which better compensates investors for the complexities that are embedded within the securities.”